CN-122025145-A - Intelligent oral health monitoring and intervention method for autism children

CN122025145ACN 122025145 ACN122025145 ACN 122025145ACN-122025145-A

Abstract

The invention relates to the technical field of big data processing and analysis, in particular to an intelligent monitoring and intervention method for oral health of autism children. The method comprises the steps of obtaining oral clinical index data, daily oral hygiene behavior record data and microbial sequencing data obtained after sequencing oral microbial samples of patients, storing the processed data in a correlated mode to form an individualized oral health file of each patient, carrying out fusion analysis through a machine learning model based on the data in the individualized oral health file, outputting current oral health state scores, risk grades and key risk factors of the patients, and further matching and generating an individualized oral health intervention scheme from a preset intervention strategy knowledge base. The method can identify whether the key factor causing risk is bacterial imbalance or insufficient behavior, improves the accuracy of etiology diagnosis, and enables intervention measures to be adaptively adjusted along with the change of oral health states of children.

Inventors

Cao Yina
Zhi Qinghui
ZUO XINRUI
YUAN ZIQING

Assignees

中山大学附属口腔医院

Dates

Publication Date: 20260512
Application Date: 20260303

Claims (8)

1. An intelligent monitoring and intervening method for the oral health of children with autism is characterized by comprising the following steps: S1, acquiring oral clinical index data, daily oral hygiene behavior record data and microbial sequencing data obtained after sequencing an oral microbial sample of a patient; s2, carrying out standardized processing on the oral clinical index data, the behavior record data and the microorganism sequencing data, and storing the processed data in a correlated way to form an individual health file of each patient; s3, based on data in the personalized health record, carrying out fusion analysis through a machine learning model, and outputting current oral health state scores, risk grades and key risk factors of the patient; and S4, matching and generating a personalized oral health intervention scheme from a preset intervention strategy knowledge base based on the oral health state score, the risk level and the key risk factor.
2. The intelligent monitoring and intervention method for the oral health of the autistic children according to claim 1, wherein in the step S3, fusion analysis is carried out through a machine learning model, and the specific steps of outputting the current oral health state score and risk level of the patient are as follows: S31, bioinformatics analysis is carried out on the microorganism sequencing data, an alpha diversity index and a beta diversity distance matrix of an oral cavity microorganism community are calculated, and relative abundance data of specified key bacteria are extracted; s32, calculating a plurality of behavior compliance quantitative indexes based on daily oral hygiene behavior record data; S33, inputting the alpha diversity index, the beta diversity distance matrix, the relative abundance data of the key bacteria genus and the behavior compliance quantification index into a trained random forest classifier, and outputting the oral health state score and the risk level by the random forest classifier.
3. The intelligent monitoring and intervention method for the oral health of autistic children of claim 2, wherein in S32, the plurality of behavior compliance quantitative indicators comprise tooth brushing frequency standard rate, single tooth brushing time period qualification rate and parental supervision time period duty ratio.
4. The method for intelligent monitoring and intervention of oral health in autistic children according to claim 3, wherein in S31, the alpha diversity index comprises a Chao1 index and a Shannon index, and the specified critical bacteria comprises at least one of Bacteroidota, veillonella and Capnocytophaga.
5. The method for intelligent monitoring and intervention of oral health in autistic children according to claim 4, wherein in S31, performing bioinformatic analysis on the microbiological sequencing data comprises: Performing sequence quality control on the microbial sequencing data so as to filter low-quality sequences, and clustering the filtered sequences into operation classification units; Performing species annotation on the operation classification unit based on a preset reference database to obtain sequence counts of each genus; Calculating the alpha diversity index and beta diversity distance matrix according to the sequence count and specifying the relative abundance of the key bacteria.
6. The intelligent monitoring and intervention method for the oral health of the autistic children according to claim 5, wherein in the step S33, the specific steps of outputting the oral health status score and the risk level are as follows: S331, analyzing a Shannon index in the alpha diversity index, a principal coordinate in the beta diversity distance matrix to obtain a first principal coordinate value, relative abundance data of at least one kind of key fungus genus and the behavior compliance quantization index to form a feature vector; s332, inputting the feature vector into a trained random forest classifier, and outputting probability distribution of the feature vector belonging to a plurality of preset health state categories by the random forest classifier; S333, determining the category with the highest probability value in the probability distribution as the oral health state classification; s334, according to a preset mapping rule, the oral health state classification is converted into a numerical oral health state score, and according to a preset threshold interval in which the oral health state score is located, the risk level is determined.
7. The intelligent monitoring and intervention method for the oral health of the autistic children according to claim 6, wherein in the step S3, the key risk factors are identified as follows: Establishing a statistical model which takes the variation of clinical indexes as dependent variables and takes the variation value of the relative abundance of at least one key fungus genus and the variation value of at least one behavior compliance quantitative index as independent variables; Calculating the contribution degree of each variable to the change of the dependent variable; Marking flora or behavior factors corresponding to independent variables with contribution degrees exceeding a preset threshold as the key risk factors; Wherein the critical risk factors are distinguished as either a dominant type of dysbacteriosis or a deficient type of behavioral compliance.
8. The method for intelligently monitoring and intervening oral health of autistic children according to claim 7, wherein in S4, the personalized oral health intervening scheme is generated specifically as follows: If the key risk factor is a dominant type of flora imbalance, the generated intervention program comprises advice for regulating oral flora, wherein the advice comprises at least one of dietary structure adjustment advice, probiotics use advice or antibacterial mouthwash use advice; if the key risk factor is insufficient in behavior compliance, the generated intervention scheme comprises behavior enhancement measures and parental supervision guidance, wherein the behavior enhancement measures comprise at least one of setting a visual tooth brushing timer, establishing a reward point system or introducing an interactive tooth brushing game.

Description

Intelligent oral health monitoring and intervention method for autism children Technical Field The invention relates to the technical field of big data processing and analysis, in particular to an intelligent monitoring and intervention method for oral health of autism children. Background Children with autism spectrum disorder have particularly prominent oral health problems due to unique perception, cognition and behavior characteristics, and dental caries and gingivitis have prevalence rate remarkably higher than that of common children, so that the children become weak links in overall health management of the children. At present, oral health intervention for the group mainly depends on oral propaganda, graphic manual or general animation video for common children, and few attempts are made to introduce a simple timer or visual prompt tool. These existing approaches are essentially static, unidirectional messaging or behavioral alerts. However, existing intervention schemes suffer from a systematic deficiency: 1. lack of true individuality, failing to adapt to sensory sensitivity, cognitive level, and microbial ecological base line of ASD children individuals; 2. Parental roles are not systematically integrated as critical supervisor and executor in the intervention closed loop, resulting in home care disjointing; 3. the intervention effect evaluation stays on the clinical observation or questionnaire level, cannot quantify the intervention effect from the intrinsic biological mechanism level of oral cavity microecology, and lacks long-term and dynamic tracking and adjusting capability. Therefore, the intelligent monitoring and intervention method for the oral health of the autism children is provided. Disclosure of Invention The invention aims to provide an intelligent monitoring and intervention method for oral health of autism children, which aims to solve the problems in the background technology. In order to achieve the above purpose, the invention aims to provide an intelligent monitoring and intervention method for oral health of an autistic child, which comprises the following steps: S1, acquiring oral clinical index data, daily oral hygiene behavior record data and microbial sequencing data obtained after sequencing an oral microbial sample of a patient; s2, carrying out standardized processing on the oral clinical index data, the behavior record data and the microorganism sequencing data, and storing the processed data in a correlated way to form an individual health file of each patient; s3, based on data in the personalized health record, carrying out fusion analysis through a machine learning model, and outputting current oral health state scores, risk grades and key risk factors of the patient; and S4, matching and generating a personalized oral health intervention scheme from a preset intervention strategy knowledge base based on the oral health state score, the risk level and the key risk factor. As a further improvement of the technical scheme, in S3, the specific steps of outputting the current oral health status score and risk level of the patient by performing fusion analysis through a machine learning model are as follows: S31, bioinformatics analysis is carried out on the microorganism sequencing data, an alpha diversity index and a beta diversity distance matrix of an oral cavity microorganism community are calculated, and relative abundance data of specified key bacteria are extracted; s32, calculating a plurality of behavior compliance quantitative indexes based on daily oral hygiene behavior record data; S33, inputting the alpha diversity index, the beta diversity distance matrix, the relative abundance data of the key bacteria genus and the behavior compliance quantification index into a trained random forest classifier, and outputting the oral health state score and the risk level by the random forest classifier. As a further improvement of the technical scheme, in S32, the plurality of behavior compliance quantization indexes include a tooth brushing frequency standard reaching rate, a single tooth brushing time period qualification rate and a parental supervision time period duty ratio. As a further improvement of the present technical solution, in S31, the α -diversity index includes a Chao1 index and a Shannon index, and the specified key genus includes at least one of Bacteroidota, veillonella and Capnocytophaga. As a further improvement of the present technical solution, in S31, performing bioinformatic analysis on the microbiological sequencing data includes: Performing sequence quality control on the microbial sequencing data so as to filter low-quality sequences, and clustering the filtered sequences into operation classification units; Performing species annotation on the operation classification unit based on a preset reference database to obtain sequence counts of each genus; Calculating the alpha diversity index and beta diversity distance matrix according to th